Detect CSV anomalies using Z-score, IQR, and moving-average methods.
npx skills add https://github.com/haomingz/kimi-skills --skill outlier-scan
Please help me install this Agent Skill. Skill: outlier-scan Source: https://github.com/haomingz/kimi-skills/tree/main/skills/outlier-scan Command: npx skills add https://github.com/haomingz/kimi-skills --skill outlier-scan
CSV data often contains outliers that distort analysis and complicate data quality assessments. This tool detects outliers using Z-score, IQR, and moving-average methods, then categorizes them for targeted review and reporting.
Run python3 scripts/anomaly_detector.py data.csv to detect outliers across numeric columns and generate a JSON report.
High-intent search queries and answers about installing and using this skill.
You can detect outliers by applying Z-score, IQR, and moving-average methods across automatically detected numeric columns in your CSV dataset. The tool generates a structured JSON report with per-row anomaly details and categorizes each anomaly as needs_attention or explainable.
Identifying anomalies in sensor streams or finance data uses multi-method detection combining Z-score, IQR, and moving-average to cover different data distributions. Each detected outlier is labeled explainable or needs_attention to prioritize investigation and remediation.
Yes, the moving-average anomaly detection method supports configurable thresholds and window sizes. You can adjust these parameters to fit your dataset characteristics and generate per-column statistics that support remediation decisions.
Generating a JSON report for CSV data quality involves running multi-method anomaly detection across numeric columns to produce per-entry method results and per-column statistics. The structured output summarizes per-row anomaly details for downstream analysis and dashboards.
Automatic numeric column detection works with mixed-type CSV datasets by identifying and processing only numeric columns for outlier detection. Non-numeric data is excluded from Z-score, IQR, and moving-average calculations to ensure accurate anomaly results.
Use IQR for anomaly detection when your data is skewed or non-normally distributed since it relies on quartiles. Use Z-score when data follows a normal distribution, and apply moving-average methods for time-ordered or sequential data to detect contextual outliers.
h4vzzClean raw datasets into analysis-ready tables using pandas, pyjanitor, and great_expectations.
kaifashraffExtract GA4 analytics data and detect anomalies via Google Analytics Data API.
brandonburrusApplies statistical methods to analyze data distributions, detect anomalies, and test hypotheses.
milasaurusDetect and explain anomalies in datasets using Isolation Forest and Local Outlier Factor.
